facebook twitter rss feeds

Wind Power Prediction Contest

Conference Program
View by day or program, includes author index and personal scheduler

The Wind Power Forecasting contest is formulated to challenge participants to forecast wind power given a list of predictors. A training dataset will be available on the website of the Artificial Intelligence Committee by June 2011(http://ai.metr.ou.edu/). Participants can use those data to train a method of their choice. The technique will then be applied to a test dataset that withholds the predictand.  Results will be judged according to previously announced metrics. The contest will include two options: 1.forecasting a long time period with a goal of matching power output, and 2. forecasting wind power ramps. Both novices and wind power forecasting professionals are invited to participate.  The identity of participants will be withheld in announcing the rankings (with the exception of the winners).
The goal of the contest is twofold: first, to challenge individuals to try new statistical learning methods and second, to elevate the state-of-the-science of wind power forecasting by sharing the results of various techniques throughout the community.

Host Conference:
10th Conf on Artificial Intelligence Applications to Environmental Science

Partnering Conference(s):
Third Conference on Weather, Climate, and the New Energy Economy and Renewable Energy Subcommittee
21st Conference on Probability and Statistics